Reply to Dr. Chandrasekar (Clin Infect Dis 2001; 32:320-1) and Drs. Marr and Boeckh (Clin Infect Dis 2001; 32:321)

Sir—The point raised by Dr. Chandra-sekar [1] regarding treatment with flu-conazole is appropriate, and we thank him for reminding us about the 1997 guidelines from the Infectious Diseases Society of America (IDSA) on treatment of fever in patients with neutropenia [2]. Although in our guidelines the " Treatment Options " subsection of the section on fever and neutropenia said that flu-conazole therapy was " often inappropriate , " it is not always inappropriate, and further elaboration on this point is useful. Not all neutropenic patients have the same risk of infection caused by Aspergillus or other filamentous fungi, and flucona-zole could be used in carefully selected circumstances. As a consequence, we would amend the guidelines to include the following Key Recommendation: Fluconazole (400 mg/day) has been used successfully in selected patients (AI) [3–5], and could be considered as an alternative strategy if (1) the patient is at low risk for infection due to Aspergillus or other filamentous fungi, (2) the patient lacks any findings that suggest that the current fever might be due to Asper-gillus or other filamentous fungi (studies should include high-quality CT of the chest plus any other clinically indicated sites), (3) local epidemiology suggests that the patient is at a low risk for infection with azole-resistant isolates of Candida species, and (4) the patient has not received an azole antifungal agent as prophylaxis. Dr. Chandrasekar also comments on the dosage of amphotericin B to be used for the treatment of aspergillosis. The comments in guidelines for the treatment of candidiasis [6] specifically address fever of unknown etiology, and the stated dosages of amphotericin B are the most commonly used dosages for patients with such fever. As discussed in the accompanying paper on guidelines for asper-gillosis [7], maximum tolerated dosages of amphotericin B are indeed appropriate if aspergillosis is proven or strongly suspected. Dr. Chandrasekar closes his comments with an aside about the timing of initiation of antifungal therapy. We recommended that the doctor consider starting antifungal therapy if a patient with fever fails to respond despite having received 4-6 days of suitable antibacterial therapy (i.e., on or after day 5 of persistent fever); this allows for sufficient time for resolution of a fever caused by a bacterial infection. It also follows the design of most major trials in this area, and is consistent with the IDSA's recommendation to start therapy during days 5–7, a recommendation that was …